Reaction of Zn+ with N02. The gas-phase thermochemistry of ZnO D. E. Clemmer, N. F. Dalleska, and P. B. Armentrouta) Department of Chemistry, University of Utah, Salt Lake City, Utah 84112 (Received 24 Iune 199 1; accepted 2 August 199 1) The homolytic bond dissociation energies of ZnO and ZnO + have been determined by using guided ion-beam mass spectrometry to measure the kinetic-energy dependence of the endothermic reactions of Zn + with nitrogen dioxide. The data are interpreted to yield the bond energy for ZnO, D g = 1.61 t 0.04 eV, a value considerably lower than previous experimental values, but in much better agreement with theoretical calculations. We also obtain D g (ZnO + ) = 1.67 k 0.05 eV, in good agreement with previous results. Other thermochemistry derived in this study is D i (Zn +-NO) = 0.79 f 0.10 eV and the ionization energies, IE( ZnO) = 9.34 + 0.02 eV and IE( NO, ) = 9.57 + 0.04 eV.
INTRODUCTION The measurement of neutral metal-ligand thermoche- mistry from the reactions of metal ions is accomplished by In a recent paper on the thermochemistry of gas-phase studying the endothermic transfer of a negatively charged transition-metal oxide neutral and ionic diatoms,’ we noted ligand to the metal, reaction ( I ) : that there is little information about the gas-phase thermo- chemistry of zinc oxide. Zinc is the only first-row transition M+ +XL-+ML+X+ (1) metal for which there have been no definitive measurements -rML+ +X. (2) made of the neutral oxide bond energy, D O(ZnO), and ioni- zation energy, IE( ZnO). In 1964, Anthrop and Searcy* at- This type of reaction has been used previously to obtain neu- tempted to study this molecule by high-temperature mass tral metal-hydride,8-‘s -methyl,9*‘0*‘2*‘6 and -oxide’*,“,‘* spectrometry, but did not detect “gaseous zinc oxide mole- bond energies. Since metal-containing species such as ML cules of any kind.” Based on estimates of their experimental have fairly low IEs, a competing process is reaction (2). sensitivity, they assigned an upper limit for the zinc oxide Successful observation of reaction ( 1) therefore requires bond energy of Di,, (ZnO) G2.86 eV. This same data was that the X fragment have a relatively low IE. The reevaluated in 1983 by Pedley and Marshall in their compi- Zn + + NO, system, where reaction ( 1) leads to formation lation on gaseous monoxides and they cite of ZnO+NO+, is chosen for the present study because DE ( ZnO) < 2.77 + 0.43 eV.3 In this same year, a lower lim- IE(N0) = 9.264 36 + 0.000 06 eV,19 much lower than the it of D ‘( ZnO) 22.8 + 0.2 eV was obtained by Wicke, who IEs of fragments of other small oxygen-containing mole- monitored the chemiluminescent reaction of zinc atoms cules such as 0, , CO,, alcohols, ketones, and aldehydes. with N, 0 as a function of kinetic energy.4 Wicke goes on to suggest that his lower limit, along with the upper limit of EXPERIMENTAL SECTION Anthrop and Searcy gives D ‘( ZnO) -2.8 eV. However, this General value is well above the theoretical values of D, = 1.20,0.66, Complete descriptions of the apparatus and experimen- and 1.44 eV calculated by Bauschlicher and Langhoff,s tal procedures are given elsewhere.20 Zn + production is de- Dolg et aZ.,6 and Igel-Mann and Stall,’ respectively. scribed below. The ions are extracted from the source, accel- Bauschlicher and Langhoff commented on this large dis- erated, and focused into a magnetic sector momentum crepancy but were unable to reconcile the difference. analyzer for mass analysis. Mass-selected ions are slowed to In this paper, we report the first direct measurement of a desired kinetic energy and focused into an octopole ion the bond energy and IE of the neutral ZnO molecule. This is guide that radially traps the ions. The octopole passes achieved by using guided ion-beam techniques to character- through a static gas cell containing the neutral reactant at ize the kinetic-energy dependence of the reaction of Zn+ pressures sufficiently low ( -0.05-0.13 mTorr) that multi- with NO,. In addition, we also measure the bond energy for ple ion-molecule collisions are improbable. After exiting the ionic zinc oxide, and compare this with a value previously gas cell, product and unreacted beam ions drift to the end of measured from reaction of Zn+ + O,, 0: (ZnO + ) the octopole where they are directed into a quadrupole mass = 1.65 f 0.12 eV.’ Further, since both the ionic and neu- filter for mass analysis and then detected. Ion intensities are tral zinc oxide molecules are formed in the present reaction converted to absolute cross sections as described previous- system, the energetic difference between these reaction path- ly.20 Absolute uncertainties in cross sections are generally ways allows a direct measurement of IE( ZnO). about + 20%. Relative uncertainties are much smaller. All product cross sections reported are the result of single ion- molecule collisions as verified by examining the pressure de- pendence of the product intensities. Laboratory ion energies are related to center-of-mass ” Camille and Henry Dreyfus Teacher-Scholar, 1987-1992. (CM) frame energiesby EC,, = Elabm/(M + m) whereM
DownloadedJ. Chem. 28 JunPhys. 2002 95 (lo),to 129.79.63.125. 15 November 1991Redistribution 0021-9606/91/227263-06$03.00 subject to AIP license or copyright,@ 1991 see American http://ojps.aip.org/jcpo/jcpcr.jsp Institute of Physics 7263 7264 Clemmer, Dalleska, and Armentrout: Reaction of Zn+ with NO,
and m are the ion and neutral reactant masses, respectively. o(E) ~00 Cgi(E-Eo +E, +Erot)“/Ey (3) Sharp features in the observed cross sections are broadened I by two effects: the thermal motion of the neutral gas, which which involves an explicit sum of the contributions of indi- has a width of -0.41E&$ for these reactions,*’ and the vidual reactant states, denoted by i, weighted by their popu- distribution of ion energies. The zero of the absolute energy lations, g,. Here, o, is a scaling factor, E is the relative kinet- scale and the ion energy distribution (full width at half maxi- ic energy, n is an adjustable parameter, and E, is the 0 K mumz0.4 eV lab) are measured by a retarding potential threshold for reaction of the lowest electronic level of the ion technique described elsewhere.*’ The uncertainty in the ab- with the lowest (O,O,O) vibrational level of NO,. E, repre- solute energy scale is + 0.05 eV lab. sents the vibrational levels of NO, populated at 305 K (the nominal temperature of the octopole) as calculated by a Ion sources Maxwell-Boltzmann distribution24*3’*32 and E,,, is the rota- Zn + used in these experiments has been produced by tional energy of the neutral reagent, which for room tem- two sources: electron-impact ionization of zinc metal vapor perature NO, is 3kT/2 = 0.039 eV. Before comparison with and argon-ion impact on a zinc metal surface. The electron- the experimental data, this model is convoluted with the impact source, described in detail previously,‘6 uses a resisti- neutral and ion kinetic-energy distributions as described vely heated oven to vaporize zinc powder. The resulting zinc previously.2o The mo, n, and E, parameters are then opti- atoms are ionized by electron impact at electron energies mized by using a nonlinear least-squares analysis to give the between 10 and 14 eV. Ionization of Zn with electrons hav- best reproduction of the data. Error limits for E. are calcu- ing energy less than 14 eV ensures that Zn + is produced in lated from the range of threshold values obtained for seven its electronic *S ground state since the ionization energy of different data sets with different values of n and the error in Zn is 9.394 eV and the first excited state of Zn + lies 6.01 eV the absolute energy scale.33 higher in energy. ** The second source is a flow-tube In order to model some data channels, we also use a source23 that uses a dc discharge24 in a 10% Ar in He flow to modified form of Eq. (3) which accounts for a decline in the sputter a zinc metal surface and create zinc ions. Both product ion cross section at higher kinetic energies. This sources yield similar results indicating that only ground- model has been described in detail previously,34 and de- state Zn + (*S) is formed. pends on ED, the energy at which a dissociation channel or a competing reaction can begin, andp, a parameter similar ton NO, purification in Eq. (3). Nitrogen dioxide as obtained from Matheson in 99.5% purity was subjected to multiple freeze-pump-thaw cycles RESULTS using liquid nitrogen. During the first freeze cycle, the solid Four products, formed in reactions (4)-( 7), phase (N, 0, > was always a blue-green color indicating that N, 0, was also present. Before thawing, an excess of oxygen Zn + (*S) + NO, NO,C + Zn (4) gas (roughly 10-20 times more than the initial NO, added) ZnO+ + NO (5) was introduced to the system. The frozen nitrogen oxides were allowed to warm up in the excess of oxygen such that NO+ + ZnO (6) any NO formed from decomposition of N,O, reacted with 1 ZnNO + + 0, oxygen to form NO,. Upon refreezing the resulting NO, (7) and oxygen mixture, the resulting solid phase was always a are observed from the reaction of Zn + (*S) with NO,. Fig- white powder with a slight yellow tint (indicating that N, 0, ure 1 shows the cross sections for these processes as a func- was the main species). The excess oxygen was pumped away tion of kinetic energy.3s All processes have cross sections while the N204 was held at liquid-nitrogen temperatures. that exhibit thresholds, suggesting that they are all endo- Pressures in the bulb were kept below - 50 Torr, in order to thermic, which immediately establishes several thermody- favor NO2 in the 2N0, F?N, 0, equilibrium. namic relationships: IE( NO, ) > IE( Zn), D ‘( NO ‘-0 - ) Results for the reaction of Zn + with unpurified NO, > D ‘(Zn + -0 - ) (the heterolytic bond dissociation ener- were drastically different than the data obtained with the gies),andD~(NO-O)>D”(Znf-O)andD’(Zn+-NO) purified gas. The main difference was the observation of (the homolytic bond dissociation energies). Analysis of the large amounts of NO + formed in an exothermic process and threshold regions of these cross sections is discussed in detail small amounts of NO*+ formed exothermically. Since below and will more precisely define these thermodynamic IE( NO) < IE( Zn), we believe that the exothermic behavior relationships. in the NO + cross section is probably due to charge transfer At low energies, the charge-transfer process, reaction of Zn + with an NO contaminant. The origins of the exother- (4), dominates the reactivity. This process is near thermo- mic formation of NO: are less clear. After purification, the neutral since the onset of the NO,+ cross section is only exothermic behavior in both the NO + and NO,+ channels slightly above 0.0 eV (Fig. 1). The cross section for reaction was negligible. (4)) a( NO,f ), reaches a maximum near 1.O eV and then declines. This behavior cannot be explained by decomposi- Thermochemical analyses tion of the NO,+ ion since this process requires at least 2.79 Theory2s,26 and experiment27-30 show that cross sec- eV=D’(NO+ -O), as calculated from the literature ther- tions for endothermic reactions can be modeled by Eq. (3 ), mochemistry in Table I. Instead, the decline in CJ(NO*+ ) is
J. Chem. Phys., Vol. 95, No. IO, 15 November 1991 Downloaded 28 Jun 2002 to 129.79.63.125. Redistribution subject to AIP license or copyright, see http://ojps.aip.org/jcpo/jcpcr.jsp Clemmer, Dalleska, and Armentrout: Reaction of Zn+ with NO, 7265
ENERGY k4, Lob) TABLE II. Parameters of Eq. (3) used to analyze reaction cross sections.”
O10 Product E. (eW 00 n ED
NO;+ 0.18(0.04) 1.8cO.3) 1.3(0.1) 1.2CO.l)
f ZnO + 1.45(0.05) 6.4( 1.4) l.O(O.1) NO+ NO+ 1.38(0.04) 10.5( 5.3) l.O(O.2) . . . t ZnNO + 2.33(0.10) l.O(O.2) l.O(O.3)
a Uncertainties in parentheses. b,. l 0 n.. -... ¤=~=m~m~mmm -m... l . . _ **0*0*** 0, on qo** 00 0 of NO+ +Zn+O cannot begin until 2.99 *(I,, ZnNO+ eV = D’(NO-0) + IE(NO)-IE(Zn) (Table I). Like- .O 0 10-l ,.I,, t, I t, t I I t, I I I I, I b wise, ZnO + formation must be accompanied by formation 0.0 1.0 2.0 3.0 4.0 ENERGY (PV. CM) of NO since production of ZnO + + N + 0 would require 8.0 + 0.1 eV (Table I). Since the ZnO + NO + and FIG. 1. Variation of product cross sections for reaction of Zn + with NO, as ZnO + f NO products differ only in the location of the a function of translational energy in the center-of-mass frame (lower scale) charge, these reaction channels must compete directly, as and the laboratory frame (upper scale). Results are shown for NO,+ (solid confirmed by the similar energy dependence of these cross squares), NO ’ (solid circles), ZnO + (open circles), and ZnNO + (open sections throughout the energy range examined. Further, squares). The solid line represents the total reaction cross section. The data the similar thresholds of reactions (5) and (6) mean that shown are the average of several independent data files. IE(Zn0) zIE(N0).
THERMOCHEMISTRY If there are no energy barriers in excess of the reaction probably due to competition with reactions (5) and (6), endothermicity, as is generally the case for endothermic ion- since the peak in a( NO,+ ) occurs near the onsets for these molecule reactions,30*36 then the observed thresholds for re- reaction channels (Fig. 1) . The total reaction cross section actions (4)-( 7) can be used to derive thermochemistry for increases substantially at the thresholds for reactions (5) the product species. This assumption can be tested in the and (6), indicating that these two reaction channels are present system since literature information on reactions (4) more facile than reaction (4) even though the latter process and ( 5 ) is available for comparison. For all four reactions, is more energetically favorable by about 1 eV. the optimized fitting parameters of Eq. (3) are given in Ta- The apparent threshold for a( NO + ) at about 1 eV indi- ble II. cates that NO + formation must be accompanied by produc- Reaction (4)) charge transfer between Zn and NO,, has tion of the neutral ZnO species since formation a thermodynamic threshold that corresponds simply to the
TABLE I. Literature thermochemistry at 0 K (eV).”
M A,H’(W IE A,H’(M+)
NO, 0.372(0.008) 9.586(0.003)b 9.958(0.008)’ 9.57(0.04)d NO 0.930(0.002) 9.264 36(O.C00 06)’ 10.195(0.002)’ N- 4.880(0.001) 0- 1.097(0.001) 1.461 122 2.558(0.001) Zn 1.346(0.002) 9.394 20(0.000 02)f 10.740(0.002)’ ZnO > 1.13(0.43)~ <1.1(0.2)” 11.65(0.12)’ 2.29(0.04)d 9.34(0.02)d ll.63(0.05)d
‘Values taken from Ref. 32 unless noted otherwise. Uncertainties in parentheses. b Reference 40. ‘CalculatedfromA~H”(M+) = AlHo +IE(M). dThis study. ‘Reference 19. ‘C. M. Brown, S. G. Tilford, and M. L. Ginter, J. Opt. Sot. Am. 65, 1404 (1975). z Reference 3. hReference 4. ’ Reference 1.
Downloaded 28 Jun 2002 to 129.79.63.125. RedistributionJ. Chem. Phys., subjectVol. 95, No.to AIP 10.15 license November or copyright, 1991 see http://ojps.aip.org/jcpo/jcpcr.jsp 7266 Clemmer, Dalleska, and Armentrout: Reaction of Zn+ with NO?
difference in ionization energies of these two species. The ENERGY
ENERGY
02 ’ Use of this equation and the literature thermochemistry in 8 Table I yields a value of D t (ZnO) = 1.6 1 + 0.04 eV. The ionization energy of ZnO can be determined by
I ’ I - ’ I 3 8 combining the thermochemistry for ZnO and ZnO + as 0.0 1.0 2.0 shown by ENERGY ‘kV. CM> IE(Zn0) = Dz(ZnO) + IE(Zn) - Dg(ZnO+). (9) FIG. 2. Cross section for formation of NO,+ as a function of kinetic energy This procedure yields IE(Zn0) = 9.33 + 0.06 eV. A more in the center-of-mass frame (lower axis) and laboratory frame (upper axis) direct and precise method of obtaining IE(Zn0) from our in the threshold region. The solid line is Eq. (3) with the parameters in data is to measure the relative thresholds for reactions Table II convoluted over the experimental kinetic-energy distribution, while the dashed line is the unconvoluted model. The arrow at 0.18 eV (5) and (6), since these are unaffected by any systematic shows the threshold for reaction (4). errors associated with determining absolute thresholds.
J. Chem. Phys., Vol. 95, No. lo,15 November 1991 Downloaded 28 Jun 2002 to 129.79.63.125. Redistribution subject to AIP license or copyright, see http://ojps.aip.org/jcpo/jcpcr.jsp Clemmer, Dalleska, and Armentrout: Reaction of Zn+ with NO* 7267
The average difference in these thresholds, AE, average of - 0.7 electrons are transferred from the zinc atom = E. (ZnO * > - E, (NO + ), is determined to be 0.07, to the oxygen atom.5 When ZnO is ionized, an antibonding f 0.02 eV. Since AE, = IE(Zn0) - IE( NO), we find that 4r electron is removed’ [this is largely an electron centered IE(Zn0) = 9.34 + 0.02 eV. on the oxygen atom but can be thought of as the promoted Analysis of the ZnNO + cross section with Eq. (3) Zn(4p) electron that was donated to 0 in the ionic bonding yields the optimized fitting parameters given in Table II. picture]. Since Zn + is a much poorer electron donor than Combining the threshold of E. = 2.33 & 0.10 eV with 0: Zn, the extent of ionic bonding in ZnO + should be substan- (ON-O) leads to D z (Zn +-NO) = 0.79 + 0.10 eV; how- tially reduced from that in ZnO; however, Zn + (4s’3d ‘O) no ever, since IE( NO) < IE(Zn) (Table I), this bond energy longer needs to promote an electron in order to form a strong may also be thought of as D i (Zn-NO + ) = 0.66 f 0.10 eV. covalent bond. These two effects apparently cancel one an- other such that the bond energies of ZnO and ZnO + are DISCUSSION similar. Theoretical calculations on this problem would be of interest. While the values for IE(N0, ) and 0: (ZnO + ) ob- In light of this discussion, it is also useful to consider the tained in this study are consistent with literature thermoche- bonding in the ZnNO + molecule. One might expect the mistry, the value obtained here for D ‘( ZnO) is considerably ZnNO + molecule to have a covalent bond formed by inter- smaller than the previous experimental determinations of action of the 4s electron on Zn + and the unpaired r* elec- approximately 2.8 eV.‘+ It is, however, in much better tron on NO. This bonding scheme predicts the Zn +-NO agreement with the theoretical D, values of 1.20,’ 0.66,6 and bond energy to be similar to D ‘( Zn +-H) = 2.36 + 0.13 eV 1.44 eV.’ These values can be converted to 0 K bond energies and D ‘(Zn +-CH, ) = 3.06 + 0.14 eV,16 since these spe- of 1.16,0.62, and 1.40 eV, respectively, by using a vibrational cies have strong single covalent bonds, as discussed pre- frequency of 646 cm - ‘.5v6 Our bond energy is somewhat viously. ‘6v43An alternative way of viewing the bonding in greater than the calculated values, but this is reasonable this molecule notes that IE(N0) < IE(Zn), such that the since Bauschlicher and Langhop note that their calcula- lowest-energy dissociation pathway for ZnNO+ is to tions should slightly underestimate the bond energy due to Zn( ‘S) + NO + ( ‘Z + ). Thus, the bonding between these basis-set saturation, correlation, and relativistic effects, and two closed-shell species may be largely electrostatic. The rel- Dolg et ~1.~ suggest that their value is substantially low due to insufficient treatment of electron correlation. It seems atively weak bond energy, D z (Zn-NO + ) = 0.66 + 0.10 highly unlikely that the calculated values differ from the true eV, suggests that the latter bonding scheme may be more bond energy by a factor of 2 or more since such calculations accurate. Theoretical calculations on such species would be have been shown to yield quite accurate results. For exam- of interest in clarifying this situation. ple, Bauschlicher and Langhoff also calculate D g ( CuO) = 2.79 eV,4’ in excellent agreement with the ex- Comparison with previous results perimental value 0: (CuO) = 2.75 + 0.22 eV.3 The CuO Assuming that our value for D ‘( ZnO) is correct, it is of result can also help verify the accuracy of the present experi- interest to reassess the previous experimental results in order mental method since preliminary results from our laborato- to explain the apparent discrepancies. Since Anthrop and ry on the reaction of Cu’ with NO, find that Searcy2 did not actually observe any zinc oxide molecules, DO,(CuO) = 2.85 + 0.15 eV.18 the present result is completely consistent with their derived To further assure that the present value is correct, we upper limit for D’(Zn0). The origins of the discrepancy can also consider the bonding in ZnO by comparing this with Wicke’s4 lower limit of 2.8 + 0.2 eV are less evident. species to other simple molecules containing zinc. The bond There are no obvious problems in the interpretation of energies for the zinc hydride and zinc methyl molecules are Wicke’s data, although his threshold analysis hinges on a well established, D’(ZnH) = 0.87 f 0.04 eV (Ref. 42) and relatively weak tail in his chemiluminescence data. If we use D’(Zn-CH3 ) = 0.84 + 0.14 eV.16 These bond energies are our bond energy for ZnO, we calculate that he should have the lowest of all first-row transition-metal hydride and observed a kinetic-energy threshold for chemiluminescence methyl molecules,9v43 a result that is easily rationalized by near 2.2 eV, In fact, this is approximately the energy where noting that the 4s and 3d orbitals of neutral Zn(433d lo) are the chemiluminescence data begins to rise above the Arrhen- already fully occupied. Therefore, bonding to Zn requires ius model used by Wicke to fit his threshold data. This im- promotion of a 4s electron to a 4p orbital. For similar rea- plies that the weak chemiluminescent tail does not corre- sons, it seems probable that the ZnO bond energy should be spond to the reaction of ground-state Zn with N, 0 to form lower than those of other first-row transition-metal oxides,’ ground-state ZnO. consistent with our value of D’(Zn0) = 1.61 +. 0.04 eV, One possible explanation for a kinetic-energy onset to but not with a value of 2.8 eV. chemiluminescence that is too low is the presence of excited- The observation that D’(Zn0) is somewhat stronger state Zn. Wicke considered this and estimated that the zinc than D’(ZnH) and D’(ZnCH, ) can be attributed to the beam used in his experiment has less than 0.1% excited-state relatively ionic nature of ZnO compared with ZnH and Zn atoms, suggesting that the reactivity observed is mainly ZnCH, . This is because oxygen has a higher electron affinity due to reaction of ground-state atoms. However, the weak than H or CH,.19 Indeed, Bauschlicher and Langhoff have tail at low kinetic energies could still be explained if the reac- calculated that at the equilibrium bond distance of ZnO, an tion efficiency of this small percentage of excited states was
Downloaded 28 Jun 2002 to 129.79.63.125. J.Redistribution Chem. Phys., Vol.subject 95, No. to AIPlo,15 licenseNovember or copyright, 1991 see http://ojps.aip.org/jcpo/jcpcr.jsp 7268 Clemmer, Dalleska, and Armentrout: Reaction of Zn+ with NO*
much higher than that of ground-state zinc. Since the W. G. Mallard, J. Phys. Chem. Ref. Data 17, Suppl. 1 ( 1988). ground state of Zn is a ‘S closed shell, while the excited ‘OK. M. Ervin and P. B. Armentrout, J. Chem. Phys. 83, 166 (1985). *’ P. J. Chantry, J. Chem. Phys. 55,2746 ( 1971). states are open-shell species, this seems quite plausible. Dif- ‘*C. Corliss and J. Sugar, J. Chem. Phys. Ref. Data 11, 135 ( 1982). ferences in electronic state reactivity of l-3 orders of magni- 23For a complete description of this source see R. H. Schultz and P. B. tude have been documented in the reactions of transition- Armentrout, Int. J. Mass Spectrom. Ion Processes 107,29 (1991). metal ions.++ z4R. H. Schultz, K. C. Crellin, and P. B. Armentrout, J. Am. Chem. Sot. (in press). “N. Aristov and P. B. Armentrout, J. Am. Chem. Sot. 108, 1806 (1986). ACKNOWLEDGMENTS 26W. J. Chesnavich and M. T. Bowers, J. Phys. Chem. 83,900 (1979). 27L. Sunderlin, N. Aristov, and P. B. Armentrout, J. Am. Chem. Sot. 109, This work is supported by the National Science Founda- 78 (1987). tion, Grant No. CHE 89 17980. We are also grateful to James UP. B. Armentrout and J. L. Beauchamp, J. Chem. Phys. 74,2819 (1981); Chou, Kevin Glaeske, and Carl Howard for advice about J. Am. Chem. Sot. 103,784 ( 198 1) . purifying and handling NO,. z9B. H. Boo and P. B. Armentrout, J. Am. Chem. Sot. 109,3549 (1987). 3oP. B. Armentrout, in Advances in Gas Phase Ion Chemistry, Vol. I, edited by N. G. Adams and L. M. Babcock (JAI, Greenwich, CT, in press). “Vibrational frequencies for NO, are 1357.8,756.8, and 1665.5 cm- ’ (Ref. 32). 32M. W. Chase, C. A. Davies, J. R. Downey, D. J. Frurip, R. A. McDonald, ’ E. R. Fisher, J. L. Elkind, D. E. Clemmer, R. Georgiadis, S. K. Loh, N. and A. N. Syverud, J. Phys. Chem. Ref. Data 14, Suppl. 1 (1985) Aristov, L. S. Sunderlin, and P. B. Armentrout, J. Chem. Phys. 93,2676 (JANAF Tables). (1990). ” In the case of &ZnNO + ), only five independent data sets were analyzed. ‘D. F. Anthrop and A. W. Searcy, J. Phys. Chem. 68,2335 (1964). Two of the seven total data sets were threshold scans that did not go to 3J. B. Pedley and E. M. Marshall, J. Phys. Chem. Ref. Data 12,967 ( 1983). high enough energies to observe ZnNO + . See the Results section. 4B. G. Wicke, J. Chem. Phys. 78,6036 ( 1983). l4 M. E. Weber, J. L. Elkind, and P. B. Armentrout, J. Chem. Phys. 84,1521 ‘C. W. Bauschlicher and S. R. Langhoff, Chem. Phys. Lett. 126, 163 (1986). (1986). “The absolute magnitudes of all products but NO + were easily reproduc- 6M. Dolg, U. Wedig, H. Stoll, and H. Preuss, J. Chem. Phys. 86, 2123 ible within a 20% uncertainty. The NO + product exhibited larger varia- (1987). tions in magnitude consistent with product collection difficulties. This ‘G. Igel-Mann and H. Stoll (to be published), as cited in Ref. 6. kind of behavior is typically exhibited by products that are formed with ‘M. A. Tolbert and J. L. Beauchamp, J. Phys. Chem. 90,5015 ( 1986). little forward momentum in the laboratory frame. 9P. B. Armentrout and R. Georgiadis, Polyhedron 7, 1573 (1988). ‘6P. B. Armentrout, in Structure/Reactivity and Thermochemistry of Ions, ‘OR Ge-orgiadis, E. R. Fisher, and P. B. Armentrout, J. Am. Chem. Sot. edited by P. Ausloos and S. G. Lias (Reidel, Dordrecht, 1987), p. 97. l;l, 4251 (1989). .” D. E. Clemmer and P. B. Armentrout (unpublished). ” E. R. Fisher and P. B. Armentrout, J. Phys. Chem. 94, 1674 ( 1990). “F. C. Fehsenfeld, E. E. Ferguson, and M. Mosesman, Chem. Phys. Lett. I* L. S. Sunderlin and P. B. Armentrout, J. Phys. Chem. 94,3589 ( 1990). 4,73 (1969). I3 R. H. Schultz and P. B. Armentrout, J. Chem. Phys. 94,2262 ( 1991). 39P. C. Killgoar, G. E. Leroi, W. A. Chupka, and J. Berkowitz, J. Chem. 14Y-M. Chen, D. E. Clemmer, and P. B. Armentrout, J. Chem. Phys. 95, Phys. 59, 1370 (1973). 1228 (1991). 40K. S. Haber, J. W. Zwanziger, F. X. Campos, R. T. Wiedmann, and E. R. “Y-M. Chen, D. E. Clemmer, and P. B. Armentrout (unpublished). Grant, Chem. Phys. Lett. 144,58 (1988). “‘R. Georgiadis and P. B. Armentrout, J. Am. Chem. Sot. 108, 2119 4’S. R. Langhoff and C. W. Bauschlicher, Chem. Phys. Lett. 124, 241 (1986). (1986). The 0: value cited here is calculated from D, = 2.83 eV and “R Georgiadis and P. B. Armentrout, Int. J. Mass Spectrom. Ion Pro- o,=606cm ‘. ceks 89,227 (1989). 41K. P. Huber and G. Herzberg, Constants of Diatomic Molecules (Van ‘*D. E. Clemmer, N. F. Dalleska, S. K. Loh, and P. B. Armentrout (unpub- Nostrand, New York, 1979). lished). 43P. B. Armentrout, ACS Symp. Ser. 428, 18 ( 1990). 19S. G . Lias, J. E. Bartmess, J. F. Liebman, J. L. Holmes, R. D. Levin, and “‘P. B. Armentrout, Annu. Rev. Phys. Chem. 41,313 ( 1990).
J. Chem. Phys., Vol. 95, No. lo,15 November 1991 Downloaded 28 Jun 2002 to 129.79.63.125. Redistribution subject to AIP license or copyright, see http://ojps.aip.org/jcpo/jcpcr.jsp